doorman ollama queuing proxy

For source, see ~arrdem/source/projects/doorman.

Local Hardware (capital) $10000.00
Cloud Spent $446.10
· OpenRouter / Vast $446.10 / $0.00
Would-be: Frontier / DeepSeek / OR $9913.40 / $117.80 / $231.00
Saved vs Frontier / DeepSeek / OR $9887.40 / $91.80 / $205.00
Capital Recouped 98.9%
Avg Daily Savings $12.40/d
Days to CapEx 2.2y

Queue

State Backend User Model Endpoint Wait ≥ 1st tok ETA
No active requests

Backends

Name Health URL Cost Class Used Free Budget In-Flight Loaded Pulling
hieroglyph (openai) healthy http://llama-swap-server:11434 0 48 GB 40 GB 96 GB 0 qwen3-embedding:latest, qwen3.6:35b, qwen3.6:35b-b —
megalith (openai) healthy http://10.0.0.20:11434 1 16 GB 6 GB 24 GB 0 unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL —
openrouter (openai) healthy https://openrouter.ai/api 100 7536 GB 0 GB 0 GB 0 471 models (catalog) —

Models

Model Status Loaded On Input Tokens Output Tokens Context (tok/s) Generate (tok/s)
amazon/nova-2-lite-v1 unloaded — 60 11 0.0 13.0
amazon/nova-lite-v1 unloaded — 20 32 0.0 37.9
amazon/nova-micro-v1 unloaded — 43 33 0.0 20.3
amazon/nova-pro-v1 unloaded — 20 32 0.0 55.4
anthracite-org/magnum-v4-72b unloaded — 55 29 0.0 12.0
anthropic/claude-fable-5 unloaded — 30 15 0.0 5.3
anthropic/claude-fable-5.1 unloaded — 29 12 0.0 6.7
anthropic/claude-haiku-4.5 unloaded — 46 26 0.0 15.7
anthropic/claude-haiku-5.5 unloaded — 29 12 0.0 10.4
anthropic/claude-opus-4.1 unloaded — 24 14 0.0 2.5
anthropic/claude-opus-4.5 unloaded — 23 13 0.0 8.6
anthropic/claude-opus-4.6 unloaded — 22 12 0.0 8.3
anthropic/claude-opus-4.7 unloaded — 30 13 0.0 11.9
anthropic/claude-opus-4.8 unloaded — 26 11 0.0 7.1
anthropic/claude-opus-5 unloaded — 25 32 0.0 10.2
anthropic/claude-opus-5.5 unloaded — 28 12 0.0 6.3
anthropic/claude-sonnet-4.5 unloaded — 23 13 0.0 8.3
anthropic/claude-sonnet-4.6 unloaded — 25 15 0.0 16.5
anthropic/claude-sonnet-5 unloaded — 26 11 0.0 6.1
anthropic/claude-sonnet-5.5 unloaded — 27,058,811 1,331,533 0.0 55.0
bytedance-seed/seed-1.6 unloaded — 100 32 0.0 24.5
bytedance-seed/seed-1.6-flash unloaded — 104 32 0.0 45.4
bytedance-seed/seed-2-1-turbo unloaded — 65 32 0.0 14.8
bytedance-seed/seed-2.0-code unloaded — 52 32 0.0 22.4
bytedance-seed/seed-2.0-lite unloaded — 55 32 0.0 16.7
bytedance-seed/seed-2.0-mini unloaded — 54 32 0.0 47.2
bytedance/ui-tars-1.5-7b unloaded — 38 14 0.0 30.1
cognitivecomputations/dolphin-mistral-24b-venice-edition unloaded — 579 17 0.0 24.6
cohere/command-a-plus unloaded — 15 32 0.0 84.8
deepseek-coder:33b unloaded — 50 32 998.0 64.7
deepseek/deepseek-v4-flash unloaded — 969,031,825 7,897,949 0.0 89.0
deepseek/deepseek-v4-pro-0813 unloaded — 259,010,461 1,852,070 0.0 71.3
gemma4:26b unloaded — 414,092 125,155 4019.4 173.5
gemma4:31b unloaded — 372,486,992 13,706,345 1759.2 145.6
gemma4:e2b unloaded — 1,702,025 3,079,434 1831.2 249.6
gemma4:e4b unloaded — 75,126,786 92,880,244 9314.9 181.5
google/gemma-3-27b-it unloaded — 30 17 0.0 1.1
google/gemma-3-4b-it unloaded — 30 16 0.0 13.3
google/gemma-4-26b-a4b-it unloaded — 33 15 0.0 20.1
google/gemma-4-31b-it unloaded — 73,947,854 400,828 0.0 12.9
google/lyria-3-pro-preview unloaded — 12 1 0.0 0.7
gpt-oss:20b unloaded — 349,324 190,433 3421.6 158.3
gryphe/mythomax-l2-13b unloaded — 56 14 0.0 15.5
ibm-granite/granite-4.2-8b unloaded — 31 32 0.0 84.9
inception/mercury-2 unloaded — 18 30 0.0 75.7
inclusionai/ling-3.0-flash unloaded — 42 32 0.0 25.5
inclusionai/ling-3.0-flash-vl unloaded — 41 32 0.0 24.3
inference-net/schematron-v2-turbo unloaded — 44 32 0.0 18.4
mancer/weaver unloaded — 128 32 0.0 29.5
meta-llama/llama-3.1-70b-instruct unloaded — 27 11 0.0 14.5
meta-llama/llama-3.1-8b-instruct unloaded — 53 12 0.0 64.7
meta-llama/llama-3.2-3b-instruct unloaded — 26 8 0.0 24.7
meta-llama/llama-3.3-70b-instruct unloaded — 49 9 0.0 30.0
meta-llama/llama-4-maverick unloaded — 26 10 0.0 27.6
meta-llama/llama-guard-4-12b unloaded — 221 3 0.0 8.6
meta/muse-glimmer-30b unloaded — 74 32 0.0 31.9
nousresearch/hermes-3-llama-3.1-405b unloaded — 25 32 0.0 26.4
nvidia/nemotron-3-ultra-550b-a55b unloaded — 48,734 10,073 0.0 115.9
nvidia/nemotron-3.5-content-safety unloaded — 483 32 0.0 88.4
openai/gpt-oss-120b unloaded — 1,944,429 55,799 0.0 216.7
qwen/qwen3-235b-a22b-thinking-2507 unloaded — 13,253 85,647 0.0 56.0
qwen/qwen3-coder unloaded — 5,009,439 17,160 0.0 58.0
qwen/qwen3-coder-30b-a3b-instruct unloaded — 5,672,997 33,805 0.0 88.9
qwen/qwen3.6-35b-a3b unloaded — 24 397 0.0 113.7
qwen3-embedding:latest loaded hieroglyph 117,753,814 0 3363.4 0.0
qwen3.6:27b unloaded — 588,046 391,176 1403.0 70.6
qwen3.6:35b loaded hieroglyph 833,391,704 130,357,318 3756.6 250.6
qwen3.6:35b-b loaded hieroglyph 0 0 0.0 0.0
qwen3.8:27b unloaded — 27,714,609 1,183,665 577.1 39.2
unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL loaded megalith 1,804,377 822,315 722.7 28.7
z-ai/glm-5.2 unloaded — 18 69 0.0 68.8
z-ai/glm-5.3 unloaded — 249,092,247 987,631 0.0 143.3
z-ai/glm-5.3-flash unloaded — 18 121 0.0 55.0
~deepseek/deepseek-v4-flash-latest unloaded — 88 15 0.0 30.7

Users

User Requests Input Tokens Output Tokens Time Cost (Frontier) Cost (DeepSeek) Cost (OpenRouter) Cost (Local)
anonymous 1,631,741 2,585,668,780 238,792,162 635.2h $18898.10 $275.70 $620.30 $428.30
joe 85,232 92,460,972 4,076,863 73.2h $564.20 $9.10 $9.50 $3.00
arrdem 16,212 31,607,157 3,009,698 27.8h $233.30 $3.40 $4.60 $1.20
arrdem (via joe) 13,737 293,787,108 4,387,107 51.3h $1578.60 $27.20 $39.80 $38.40
rubberstamp (via joe) 4,825 13,184,703 1,016,041 15.4h $91.30 $1.40 $1.90 $0.70
marge (via joe) 441 2,862,167 73,122 2.2h $16.10 $0.30 $0.40 $0.10
localscene 393 2,544,675 4,046,269 12.7h $113.90 $1.00 $0.60 $0.50
dewey 88 0 0 0s $0.00 $0.00 $0.00 $0.00
drain-test 13 175 10,557 2.8m $0.30 $0.00 $0.00 $0.00
burnin 11 850 2,284 38s $0.10 $0.00 $0.00 $0.00
synthetic-validate 2 39 32 1s $0.00 $0.00 $0.00 $0.00
probe 1 0 0 0s $0.00 $0.00 $0.00 $0.00
embed-test 1 10 0 0s $0.00 $0.00 $0.00 $0.00
spill-test 1 49,770 2 7s $0.20 $0.00 $0.00 $0.00
Total 1,752,698 3,022,166,406 255,414,137 817.9h $21496.20 $318.00 $677.10 $472.20

Daily

Day (UTC) Requests Input Tokens Output Tokens Cost (Frontier) Cost (DeepSeek) Cost (OpenRouter) Cost (Local) Saved
2026-10-18 15,687 1,061,229 127,524 $7.50 $0.10 $0.20 $0.00 $7.50
2026-10-17 16,045 1,466,282 149,656 $9.40 $0.10 $0.20 $0.00 $9.30
2026-10-16 16,954 13,808,146 314,601 $25.10 $0.50 $1.10 $0.00 $25.00
2026-10-15 16,216 10,363,317 238,841 $16.60 $0.30 $0.80 $0.00 $16.50
2026-10-14 16,347 3,895,145 155,658 $10.70 $0.20 $0.30 $0.00 $10.70
2026-10-13 15,767 1,297,740 115,598 $7.60 $0.10 $0.20 $0.00 $7.60
2026-10-12 15,940 1,319,676 113,226 $7.60 $0.10 $0.20 $0.00 $7.60